Study of Dross Formation in Hot Dip Galvanizing Line Using Machine Learning

A gradient-boosting machine learning model is presented. It is designed to predict the formation of defects (drosses) on galvanized steel sheets used in the automotive industry. The influence of process parameters on the formation of defects is analyzed. This made it possible to identify the key factors affecting the quality of the coating: coil rolling speed, skin pass mill elongation and force, top zinc coating thickness, temperature in the galvanizing pot, and temperature in the furnace snout. A digital decision support system for real-time coil evaluation is implemented. It provides a simulation of the decision-making and helps to promptly manage the technological process.